I have been battling this for weeks; but that duration of over 2.5 seconds is causing visitors to drop out at times and the culprit all points back to both Woocommerce and Litespeed. The server is quite beefy with loads of ram, Object cache, PHP cache, Litespeed Enterprise and a Pro cloudflare package. On that end everything is set to fully cache, on site there's lots of tweak(s) involved but i can't walk around on the ridiculous amount of database queries executes upon things like show cart, add to cart or just opening up the WP admin page.
Things i've already done:
- HPOS activated and cut ties of the old system
- Disabled Heartbeat and manual set to 120 seconds
- Database optimized as much as i could, 800MB of before 1200MB
- Stripped as much as possible plugins i did not need or could done better
- Optimized Litespeed Enterprise
- Assigned a over 1600% speed in Cloudlinux, which means, full 16 cores.
- action scheduler stripped to a maximum of 20k rows instead of hundred of thousands.
Still performance is quite, bad knowing if i build a manual shop things load up instant. With Woocommerce it's always one big hogging pile of shit. The amount of products is around 500+ and the amount of orders is within the ten's of thousands.
I understand that, a server with a load of cores (64) but low clocked is no better then a server with less cores and higher clocked. The many cores are beneficial to other scenarios but multitasking through database queries will always be a issue. Is there anything else i can do to lower the execution time of the pages?
Edit:
Shaved off 1 second in Wp admin by disabling the check on updates and moving this into a manual update. Yes i understand the risks but i also have a (active) license on Imunify360 and using CF preset WAF tools. The plugin's name for future reference is "Disable All WordPress Updates".
I’ve heard good things of this plugin to enable better indexing in the database, maybe worth a try: [https://wordpress.org/plugins/index-wp-mysql-for-speed/](https://wordpress.org/plugins/index-wp-mysql-for-speed/)
You have a pretty small database. It makes sense to install Query Monitor and check which requests take so much time. Also, the 1 second for LiteSpeed Cache is a big number.
It makes sense to switch to Redis object caching and check the Cloudflare configuration. Maybe the site is heavily loaded by Facebook and other AI garbage bots.
> and the culprit all points back to both Woocommerce and Litespeed
I would re-investigate what pointed you to these two. I’m a data engineer, we work with TBs of hourly data and billions of rows, your database is tiny – so tiny that it shouldn’t ever present a problem unless you’re doing a cartesian join on all of them for some inexplicable reason (which those plugins won’t do, they never need to).
So I would check your method of diagnosis, and figure out how to know for sure which exact lines of code are causing the problem. Everything else is a red herring.